This project addresses stastical problems generated from collaboration with scientists in other program areas and general statistical problems of current interest. This project is a continuing activity of the Section on Mathematical Statistics. Work has been published in the following statistical areas: monitoring patient recruitment in clinical studies, sample size determination, tests of trends in categorical data, early stopping of clinical trials, sequential analysis, epidemiologic surveys, robust selection, and modeling of covariate dependence of binary sequences. Other work in progress includes space-time clustering of disease, modeling of residual drug effects, and the influence of missing data on variable selection.